Scholarlink

This study investigated the possible use of Near Infrared (NIR) spectroscopy in combination with Partial Least Square Regression (PLS-R) to build a prediction model of density, Modulus of Rupture (MOR), and Modulus of Elasticity (MOE) of 5-years old Platinum teakwood. The results showed that the model for density and MOR is adequate to predict the value in the future, but the MOE prediction model is relatively fair. Root Means Square Error validation (RMSEv) and coefficient determination of validation (R2 validation) were 0.028, and 0.77 for density, 149.41 and 0.71 for MOR, and 2.01 × 104 and 0.53 for MOE, respectively. The bands at (1) 7315 cm−1 cellulose; (2) 6281 cm−1 crystalline bonds in cellulose; (3) 5980 cm−1 lignin; (4) 5795 cm−1 lignin or C–H stretching of CH2-groups; (5) 4686 cm−1 lignin and extractive; (6) 4288 cm−1 hemicellulose and xylans were the main chemical affects the density. The peaks at (1) 7000 cm−1 cellulose/amorphous region in cellulose; (2) 5980 cm−1 lignin; (3) 5795 cm−1 lignin; (4) 5464 cm−1 semi-crystalline/crystalline region in cellulose; (5) 4195 cm−1 Lignin were the main factor for MOR. While cellulose is the main factor for the MOE prediction models. There were: (1) 6281 cm−1 crystalline bonds in cellulose; (2) 5800 cm−1 furanose/pyranose due to hemicellulose; (3) 5618 cm−1 cellulose; (4) 4808 cm−1 semi-crystalline or crystalline region in cellulose; (5) 4795 cm−1 cellulose.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Physical and Mechanical Properties of 5-years Old Platinum Teakwood by Near Infrared Spectroscopy

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Danang Sudarwoko Adi,
  • Setiawan Khoirul Himmi,
  • Wahyu Dwianto

摘要

This study investigated the possible use of Near Infrared (NIR) spectroscopy in combination with Partial Least Square Regression (PLS-R) to build a prediction model of density, Modulus of Rupture (MOR), and Modulus of Elasticity (MOE) of 5-years old Platinum teakwood. The results showed that the model for density and MOR is adequate to predict the value in the future, but the MOE prediction model is relatively fair. Root Means Square Error validation (RMSEv) and coefficient determination of validation (R2 validation) were 0.028, and 0.77 for density, 149.41 and 0.71 for MOR, and 2.01 × 104 and 0.53 for MOE, respectively. The bands at (1) 7315 cm−1 cellulose; (2) 6281 cm−1 crystalline bonds in cellulose; (3) 5980 cm−1 lignin; (4) 5795 cm−1 lignin or C–H stretching of CH2-groups; (5) 4686 cm−1 lignin and extractive; (6) 4288 cm−1 hemicellulose and xylans were the main chemical affects the density. The peaks at (1) 7000 cm−1 cellulose/amorphous region in cellulose; (2) 5980 cm−1 lignin; (3) 5795 cm−1 lignin; (4) 5464 cm−1 semi-crystalline/crystalline region in cellulose; (5) 4195 cm−1 Lignin were the main factor for MOR. While cellulose is the main factor for the MOE prediction models. There were: (1) 6281 cm−1 crystalline bonds in cellulose; (2) 5800 cm−1 furanose/pyranose due to hemicellulose; (3) 5618 cm−1 cellulose; (4) 4808 cm−1 semi-crystalline or crystalline region in cellulose; (5) 4795 cm−1 cellulose.